64 lines
2.5 KiB
Python
64 lines
2.5 KiB
Python
from PIL import Image
|
|
import os
|
|
import numpy as np
|
|
import torch
|
|
import folder_paths
|
|
|
|
|
|
def open_image_from_inputdir(filename):
|
|
input_dir = folder_paths.get_input_directory()
|
|
img_path = os.path.join(input_dir, filename)
|
|
img = Image.open(img_path)
|
|
return img
|
|
|
|
def _process_layer(layer):
|
|
img = open_image_from_inputdir(layer['image']).convert('RGBA')
|
|
img = img.resize((layer['width'], layer['height']),Image.LANCZOS)
|
|
img = img.rotate(layer['rotation'], expand=True)
|
|
return img
|
|
|
|
def _paste_image(img, position,canvas):
|
|
full_size_img = Image.new('RGBA', canvas.size, (0, 0, 0, 0))
|
|
full_size_img.paste(img, position, img)
|
|
return full_size_img
|
|
|
|
def fuse_layer(layerInfoArray):
|
|
|
|
canvas = Image.new('RGBA', (layerInfoArray['width'], layerInfoArray['height']), (0, 0, 0, 0))
|
|
type_product_layers = Image.new('RGBA', (layerInfoArray['width'], layerInfoArray['height']), (0, 0, 0, 0))
|
|
|
|
for layer in layerInfoArray['layers']:
|
|
img = _process_layer(layer)
|
|
position = (layer['position_x'], layer['position_y'])
|
|
full_size_img = _paste_image(img, position,canvas)
|
|
# 合并到主画布
|
|
canvas = Image.alpha_composite(canvas, full_size_img)
|
|
if layer['type'] == 'Product':
|
|
# 对于 type=Product 的图层，我们将其合并到 type_product_layers
|
|
type_product_layers = Image.alpha_composite(type_product_layers, full_size_img)
|
|
else:
|
|
# 从 type_product_layers 中减去当前图层遮盖的部分
|
|
mask = full_size_img.split()[3] # 使用当前图层的 alpha 通道作为蒙版
|
|
inverted_mask = Image.eval(mask, lambda x: 255 - x)
|
|
|
|
# 创建一个新的 Image 对象，包含 type_product_layers 中未被当前图层遮盖的部分
|
|
unmasked_type_product = Image.new('RGBA', canvas.size, (0, 0, 0, 0))
|
|
unmasked_type_product.paste(type_product_layers, (0, 0), inverted_mask)
|
|
|
|
# 更新 type_product_layers
|
|
type_product_layers = unmasked_type_product
|
|
return (canvas, type_product_layers,)
|
|
|
|
def pilimage_to_tensor(image):
|
|
if not isinstance(image, Image.Image):
|
|
raise ValueError("`image` must be a PIL Image object.")
|
|
|
|
# 转换为RGB并转为numpy数组
|
|
image_array = np.array(image.convert("RGB"))
|
|
|
|
# 转换为torch tensor，归一化到[0, 1]范围
|
|
image_tensor = torch.from_numpy(image_array).float().div(255)
|
|
|
|
# 添加batch维度
|
|
image_tensor = image_tensor.unsqueeze(0)
|
|
return image_tensor |